LGAIJul 2, 2024

SiamTST: A Novel Representation Learning Framework for Enhanced Multivariate Time Series Forecasting applied to Telco Networks

arXiv:2407.02258v11 citationsh-index: 1Has Code
Originality Incremental advance
AI Analysis

This work addresses forecasting challenges in telecommunication networks, presenting an incremental improvement with a novel hybrid method.

The paper tackles multivariate time series forecasting by introducing SiamTST, a framework that integrates a Siamese network with attention and other techniques, achieving significant accuracy improvements on a real-world telecommunication dataset.

We introduce SiamTST, a novel representation learning framework for multivariate time series. SiamTST integrates a Siamese network with attention, channel-independent patching, and normalization techniques to achieve superior performance. Evaluated on a real-world industrial telecommunication dataset, SiamTST demonstrates significant improvements in forecasting accuracy over existing methods. Notably, a simple linear network also shows competitive performance, achieving the second-best results, just behind SiamTST. The code is available at https://github.com/simenkristoff/SiamTST.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes